A Case Study on Development Incentives in Lane County
Melinda Rowan and Jennifer Witt
(Undergraduate students at the
Department of Economics, University of Oregon,
under the supervision of Prof. Bruce Blonigen)
July 11, 2003
While much has been written and said about Hynix Semiconductor America’s presence in the
Eugene-Springfield area, we do not know of any formal study examining its economic impact
since its location. In response, this study examines the economic impact of Hynix’s presence in
the Eugene-Springfield area over the first five years of operation, 1997 through 2001. The
following table is a summary of our study's findings.
Table A: Economic Impacts Relating to Hynix's move into Eugene, Oregon.
Negative Economic Impacts Positive Economic Impacts
Tax breaks awarded to Hynix $46.0 million Local taxes paid by Hynix $5.7 million
Systems development charge
City road enhancement $3.0 million paid by Hynix to city of Eugene $3.7 million
Chance of increased local water rates Electrical substation, paid for
if severe drought uncalculated by Hynix $8.6 million
Congestion uncalculated Facility construction of Hynix Local effects uncalculated
Local spending generated by
wages and compensation of
Hynix employees $254.5 million
Taxes paid by employees of
Hynix $2.5 million - $5.2 million
Total Negative Impact $49.0 million Total Positive Impact $275.0 million - $277.7 million
NET IMPACT: $226.0 million - $228.7 million
Other findings of the study include:
• Hynix has employed an average of 644 employees between 1997 and 2001.
• Hynix has paid a total of $5,710,616 in taxes as of 2002.
• Hynix has received a total of $46 million dollars in tax exemptions as of 2002.
• Hynix destroyed 10.4 acres of wetland to build its factory, and enhanced or restored
25.59 acres to compensate.
• EWEB designed special water and electricity rates for Hynix, which included
maintenance and expansion costs, and used independent energy contracts to insulate
Eugene customers from rate increases.
• Hynix no longer qualifies for property tax exemptions on existing buildings and property
as of 2002. It can continue to receive property tax exemptions on new investments, as
long as it increases its employment by 10% each time it applies for an exemption.
In 1994, Hyundai's semiconductor division (which later became known as Hynix) began
searching for a new factory site. In 1995, they decided to build in Eugene's enterprise zone1. A
variety of reasons can be cited for this move, ranging from jumping international tariffs to taking
advantage of the tax exemptions offered by the city of Eugene as a lure for business
development. For the purposes of this study, we will examine the business incentives Eugene and
Lane County offered Hynix and whether the cost of those incentives was compensated by the
benefits the community reaped from Hynix.
Our research will examine the development incentives that were used to entice Hynix to
move into the local area and continue to keep it here. It is important to research these incentives,
which can include free or lower cost development of roads, free land use, low cost energy
packages, and various tax exemptions, in order to find if the city regains its investment from the
company. We accomplish this through formal cost benefit analysis. If these incentives are
costing the local government more than the ultimate benefit yielded to the community by the
firm, then the incentive programs should be reexamined. Much of the cost benefit analysis done
on incentive programs has revealed that there are no guarantees of net benefits. As stated in
Evaluating Business Development Incentives, "In general, academic literature is divided between
studies finding majors effects associated with growth and those finding negative or inconclusive
results. Thus far, negative and inconclusive study findings are more numerous than those finding
positive correlations."2 Therefore it is imperative that we determine if our local development
incentives program follows or defies this trend, in order to ensure that any future development
will be beneficial to the community. The evidence we find can then be applied to plans that will
affect future business investment in Lane County.
It is important to acknowledge this study's limitations. This is not a comparison between
two kinds of development programs – for example we cannot tell whether large scale economic
development such as Hynix is better than small business development. We can only show our
findings as to the financial costs to the City of Eugene in bringing Hynix to West Eugene and the
financial repercussions to date. This study makes no attempt to forecast any future revenues, only
to report what has actually come to pass.
A Short Summary of Hynix in Eugene
Several problems arose when Eugene's population heard that a large corporation planned
to build a factory in west Eugene. The first problem was that Eugene's enterprise zone, the area
where Hynix would have to build in order to receive tax exemptions, was a semi remote wetland
with sparsely used roads leading in and out and little other development. No large scale sewage
or electrical facilities were near by, and there was wildlife living where the plant would stand. Of
the many hurdles Hynix was to face, public outcry came first. According to a Register Guard
article from June of 1995, many felt that the public had no say in the establishment of Hynix.
Then director of the Eugene-Springfield Metropolitan Partnership stated, "Anyone opposed to
large industrial projects at that site [West Eugene's Enterprise Zone] should have voiced
concerns 13 years ago  when the property's light-industrial zoning was cemented into a
metro-wide land use plan. Or, they could have spoken up 8 years ago  when the City
Council designated the area an enterprise zone, awarding property tax relief to new or expanding
As a result of these laws, Hynix was able to break ground in 1995, finish construction in
1996, and begin hiring in 1997, despite public dismay. Other obstacles for the company included
wetland restoration, electrical and sewage construction, road difficulties, and finally, the various
applications to receive the enterprise zone tax exemptions. Estimates from the County Tax
Assessor conclude that, to the date of this publication, Hynix has received some 46 million
dollars in tax exemptions. These will be discussed at length in the "cost" section of this study.
To our knowledge, a formal cost benefit analysis of the Hynix program has never been
For the context of this study, it is imperative to explain the meaning of an enterprise zone,
and what are the specific requirements of the West Eugene enterprise zone. Firstly, an enterprise
zone is any area designated by a city, county, or state, to be exclusively cultivated for business
development. How an area "cultivates" this zone can vary regionally. Some states allow direct
cash donation to be made to companies that agree to reside within their enterprise zone. Oregon
does not allow that, the chosen method of enticing business development here is to award
property tax exemptions. Of course, there are stipulations the company must follow as well.
Some must agree to reside inside the zone for a certain amount of years, employ a certain amount
of people, or pay a particular base wage.
The West Eugene Enterprise zone offered (the program had since been retired, although
companies who entered before the sunset year still qualify for incentives) a three year property
tax exemption. They can request a two year extension of the initial exemption, which is
dependent on approval from the city. The various taxes included under the classification of
"property tax" will be discussed later. In return, the corporation moving into the enterprise zone
had to meet the following requirements: the company must reside within the boundary, be non-
retail, increase employment by 10% each time it makes a new investment for which it wants a
personal property or regular property tax break, and maintain a suitable number of employees
based on a fraction of its highest employment levels. Hynix has met these requirements save one
time. When reapplying for a personal property tax break in 2002, they dipped below the 10%
employment expansion criteria. They were only awarded a partial tax exemption, which is
discussed further in the analysis section of this study. Hynix did pay an estimated $3.3 million
for the property within the zone; it was not awarded by the city.
In order to properly prepare for this study, we examined several similar papers and
consulted methodology texts to ensure that we have some sense of how to conduct our data
collection and analysis. We found many important methods for calculating and compiling data
that we chose to use.
The first component of our analysis was to find a way to calculate total economic impact
of wages and salaries. In researching, we discovered that the most common tool used for this is a
multiplier. We will more thoroughly discuss multipliers in the methodology and benefits sections
of this paper. The Economic Significance of Toyota Motor Manufacturing, U.S.A., INC, In
Kentucky, a study done by Calantone and Loof at the University of Kentucky, used multipliers to
calculate the entire impact of Toyota's wages and salaries on the state economy4. In addition, A
Study of the Economic Impact of the University of Oregon by Larry Singell, Ph.d, had extensive
advice on how to develop and use multipliers to estimate economic impacts on the Oregon State
economy. Both studies examined how an individual firm would affect local economic activity,
and develop a basis for suppliers to move into their areas.5
When deciding to use a multiplier, one must either find a pre-constructed one that is
reliable, or create one. Creating a multiplier requires complete revenue data from the individual
firm, and survey information from other local businesses to establish how much spending is
generating from the firm in question. This was beyond the scope of this study, so following the
models in the Toyota Study, as well as The Economic Impacts of a Steel Mill in Coos County by
Arthur Ayre, we choose to use an input-out model6.
The benefit of this type of impact model is that it takes into account several different
facets of economic impact, including how many suppliers will work with the firm in question,
how much revenue the firm in question will generate in the new area, multipliers for
employment, economic impact, and other useful statistics. IMPLAN is so precise, that it will
even calculate how many new people may move into the area to work for the new business, or
how many new people will be employed in other industries, such as food or services industries,
to accommodate the increasing revenue from the new firm. The Coos Bay study used IMPLAN
to calculate many of these things, although that study was completely hypothetical (how would a
new firm affect the economy) whereas our study is based on historical data. IMPLAN derives its
statistics from United States Census Bureau information. The Toyota Study used a different
input-output model. After discovering the Oregon State University maintained an IMPLAN
model for Oregon, we chose to use it, because we had access to Bruce Sorte, the individual who
maintains the model.
Finally, estimating data when it is not available can raise certain issues. When estimating
the tax base changes due to the taxes paid by Hynix's employees, we followed the method of
using averages calculated state organization such as the Oregon Department of Revenue. This
method is used in Larry Singell's study, and is a common trick for giving an estimate that, while
it may not be perfect, reflects fiscal chances with some accuracy.
Data Collection and Estimation
This study required different methodologies based on its different stages: collection
gathering, data estimation, and data analysis. We'll discuss each separately and in detail:
Ascertaining the value of costs and benefits of Hynix's move into Eugene, Oregon was
not an easy matter. Firstly, there was no single data source. The following organizations had to
be contacted to compile the information you see in this study: The City of Eugene Planning
Department, EWEB (Eugene Water and Electric Board), Hynix Corporation, The Wetlands
Mitigation Bank, The County Tax Assessors Office, The Register-Guard, Eugene's Land Use
Permit Office, Eugene's Permit Building Office, and finally, Oregon State University's
Economics Department, which kindly aided us with our IMPLAN model multipliers. In addition,
the cost of conducting city wide records searches is extremely high. Therefore we have relied
heavily on information gained through interviews, and the individuals who provided us with that
information are cited as references for this paper.
Data estimation was another challenge. Much of the information required to complete
this study was either classified (e.g. Hynix's utility bills, exact wages vs. total compensation) or
incomplete (e.g. how much money was put into the local economy when Hynix built its plant).
When possible, we estimated data. For example, when estimating taxes paid by Hynix
employees, we used state or Lane County averages to estimate the possible tax benefits. This is a
common methodology in economics and although it may seem inexact, it provides a more
complete estimate then would disregarding the unknown data entirely.
Finally, data analysis provided its own set of problems. For example, once we knew how
much Hynix paid in wages, we had to apply a multiplier to it to discover how much spending
was generated in our local community. We used a multiplier from the IMPLAN model7. The
IMPLAN model was originally developed by the University of Minnesota for the United States
Government. IMPLAN stands for IMpact PLANning, and was first heavily utilized by the
United States Forest Service. This model compiles a massive quantity of information, mostly
based off federal census data, and uses it to calculate each industry's effect on other industries.
By taking census data, it finds the links between the industries (for example, if a semiconductor
plant moves into an area, will it create suppliers? How many new residents will it attract to the
area? How much money will these people spend in the local economy versus somewhere else?)
The unique quality that IMPLAN possesses is that it uses various multipliers to
demonstrate three main things: the impact on all local firms that could become suppliers to the
new business, the impact on the local retail economy of the firm's new employees, and the
continuing multiplier effects on the local economy. A multiplier is a generated number that
provides not only the direct effect a company has on an economy (for example, the new firm will
pay an employee $34,000), but also how that direct effect is continually spent in the local
economy through input-output relationships between sectors. This is more thoroughly explained
in the benefits sections of this study, when it is actually applied.
Originally, we intended to create only a cost benefit analysis. However, due to the
difficulty of defining what may be a benefit to the economy, we chose to present data in terms of
negative and positive economic impacts. A negative impact would take money out of the
community, whereas a positive impact would bring money into it. A positive impact and a
benefit are not the same thing. For example, the wages and benefits paid to Hynix employees
create a positive impact on our local economy, but they might not entirely be a benefit. A benefit
would mean those wages and benefits would not have already been paid if Hynix were not here
(some of Hynix's workers may have been previously employed in Lane County, and would have
been paying taxes and receiving wages already). Thus, the added benefit of working for Hynix
would only be the increase in wages and taxes paid over the previous employer. This is
significantly more difficult to calculate, and we took much more liberty with those estimates.
Thus, we hope in providing you with both, and you can fully realize the Hynix's impact and
benefit to the community.
Costs and Negative Economic Impacts
We have divided the cost of Hynix's move into the Eugene area into four categories:
Taxes, Roads, Utilities, and Wetlands. We have included some things in this section that actually
turn out to be benefits because they were originally anticipated as costs. Some of these categories
imply indirect costs, i.e. if no taxes are paid, how does that affect schools? These will be
discussed as needed within each subsection.
The taxes that are lost to the community in which the firm locates can be considered a
cost or excluded from costs. There are two ways to approach this decision. The first approach is
to consider the lost tax revenue as a cost to the community because it is money they should be
collecting from the firm. Presumably, such revenues are necessary to cover the costs of local
public services and infrastructure used by the firm. When a particular business receives tax
exemptions, it is common to wonder why that particular company, and not another, should
receive special treatment.
Alternatively, a second approach is to consider the lost tax revenue as money that would
have never been generated if the firm had not located in the region. If sufficient incentives were
not offered to convince Hynix to move into the area, it begs the question that other business of its
scale may have made similar choices. In addition, not offering development incentives puts Lane
County at a disadvantage in comparison to other regions, some of which offer outright cash
awards for business development. We must consider that there is little evidence that if Hynix had
not located here that another firm would have generated the same tax revenue. For this study, we
have chosen not to take a position, but to offer this information as a guide, as well as the scenario
below, recognizing that this issue has no one right answer and is currently debated in the world
of economics. In our tables, we place the tax exemptions under negative economic impacts, to
display the most conservative number possible.
The following is a standard way that economists think about firm location decisions.
Regions competing for business development enter a situation that economists call a prisoner's
dilemma: where both parties would be better off if they communicated, but because they do not
want to reveal their positions to each other, an inefficient result ensues. Suppose there are two
regions, A and B, and they are competing for a particular firm. Suppose the firm inherently
prefers region A, because it knows it will have higher profitability there, (e.g. 4 million in profit
per year, whereas in region B they would only have 2 million in profit per year) and all other
factors are equal (both regions can offer the same amount of development incentives, suppose up
to 5 million). Each region can only bid incentives up to the point where their net benefits from
the investment would be zero. In practice, this may be difficult to do because the benefits and
costs of such incentives and firm location are unknown and must be estimated. If region B offers
5 million dollars in development incentives, it will boost profits up to 6 million a year. But
region A has the profit advantage, and thus if they offer 2.1 million in incentives, the firm's
profits will increase to 6.1 million. Therefore regions with the higher net benefit would like to
offer incentives just above the competitors offer in order to win the bidding process. In fact, in
the process, the firm receives benefits that it would not receive without the bidding war.
Some might ask why region A would bid at all. If they don't, they lose the opportunity to
have any revenue whatsoever because region B will attract the business. If the bidding process
was banned, region A would win anyway, and everyone would be better off, but the competition
is allowed, and thus region A must play the game to win the firm. In the end, the bidding process
does not change the location decision; it just transfers funds from the winning region to the firm,
unless one region decides not to play.
If one decides to include taxes as a cost, it is necessary to differentiate between the types
of taxes Hynix would hypothetically pay. The first is property tax that is not exempt. There are
taxes on the actual land, which are not exempt. This figure is small and has never been part of
the tax exemptions offered by the City of Eugene. The second is regular property tax - the
traditional tax paid on buildings and structures by any person or corporation owning property.
This tax was exempted for the first 3 years the plant existed. Hynix received a 3 year exemption
for locating into the enterprise zone, and opted to reapply and extend the exemption for two more
years thereafter, which was denied by the city. However, in 2002 the original exemption expired
and it cannot be reinstated8. As of 2002, Hynix will pay standard property tax, and has done so
since then in numbers totaling $5.2 million9.
The final kind of tax is perhaps the most controversial. Personal property tax is a figure
paid to the city, county, and state based on the value of the equipment owned by a corporation.
Hynix's equipment could very well be valued at more than its property and plant put together.
However, new equipment purchased for any corporation in the enterprise zone is eligible for a
three year tax exemption from the date of its purchase. Since the plant's construction, Hynix has
retooled before any equipment tax exemption could expire. This is a common occurrence in the
semi-conductor industry, where changing technology requires the purchase of new machinery
every 2-3 years. However the result is that Hynix has never paid personal property tax on
exempted items (there are some non-exempt items, such as office supplies, computers, and all
other “rolling stock”), and as long as it retools all of its equipment every three years, it never
An indirect cost of non-taxation is how the lack of new taxes will affect school, city, and
county budgets. The tax exemption law states that in order for a locality to disregard these
effects, the corporation will have to employ a certain number of people at all times, assuming
that the financial benefit to the community with that many more people working will cancel out
any negative tax cost. Hynix has been held to this standard. In 2002, Hynix's employment dipped
below the mandatory 986 workers it needed to qualify for a tax extension. At the time, only 747
people were employed at the factory. Due to their failure to meet the requirement, the city gave
only a partial net tax exemption. Hynix had to pay $346,405 directly to 4J, Bethel, Lane
Community College, and Lane ESD school programs, $138,855 to the city of Eugene, and
$25,356 to Lane County, thus the net tax exemption during this period was $514,107.iv
According to the city planner, this gave more revenue to schools in the long run, because if the
money had been paid to Salem through taxes, schools would have received less due to
subsequent budget cuts.
One particularly interesting finding is that misinformation permeates Eugene with
concern to Hynix's tax exemptions. When interviewing people, we found the majority believed
Hynix received a much more generous exemption than the reality. The Register Guard originally
estimated that, based off a three phase plan introduced by Hynix when construction was still in
the theoretical phase, the plant would eventually qualify for up to 170 million dollars in tax
exemptions. Many people believed that Hynix, as of 2002, had received somewhere close to this
figure. In reality, Hynix has received 46 million dollars in tax relief as of 200310. This has been
given out through 2 different sets of tax exemptions: the original property tax relief awarded
through the enterprise zone, and another exemption on just personal property obtained through
retooling in 2002.
Taxes Avoided: $46 million
Taxes Paid: $510,616 (property taxes paid for lack of employment)
$5.2 paid in property taxes
The road infrastructure in the West Eugene Enterprise Zone was not sufficient to
accommodate a large factory like Hynix when construction was being contemplated in 1995. In
order to facilitate development, the City of Eugene improved a currently existing road near the
plant. According to Denny Braud, a planner with the city of Eugene, an estimated 3 million
dollarsiv was spent to further develop the intersection of Highway 126 and Willow Creek Road,
including the installation of a traffic light and a turn lane. He states that these changes were in
the city's long term plan, and while Hynix's arrival pushed them forward, they were not
It's important to note that when a corporation moves into a city, it is required to pay a
"Systems Development Charge" or SDC, to recover costs like road impact. Hynix paid 3.711
million dollars in SDCs to the city of Eugene in accordance with this rule. For the purposes of
this study, we will count that as a benefit.
Total Cost: 3 million
Total Benefit: $3.7 million
Electricity and Sewage Development
As a semiconductor plant, Hynix requires large quantities of electricity and water to
create its product. This was a major concern to the city and Eugene's citizens. First, we will
discuss electricity effects. In 1996, EWEB (Eugene Water and Electric Board) and Hynix
directly created a power sales agreement stipulating that Hynix would pay up front for the new
electrical substation that needed to be built to accommodate the plant. However, that substation
is owned by the city of Eugene and EWEB, so this could actually be seen as a benefit. The
construction of the plant cost $8.6 million dollars. EWEB made a short term loan to Hynix, for
which its then parent company, Hyundai, signed a guarantor notice. The loan was paid to EWEB
within the three years and without delays.
EWEB also realized there would be a local concern that Hynix's power usage would raise
the electricity rates for other local customers. In order to prevent this, they purchase independent
energy contracts to cover all of Hynix's electricity needs. This, in essence, shields the rest of the
system from Hynix's energy use. The cost of buying these independent contracts, as well as the
cost of any maintenance needed for Hynix's electricity station, is built into the rate that they pay.
EWEB representative Mel Taylor states that EWEB made the conscience choice to charge Hynix
for all of these things in order to not raise energy rates for the community. He also stated that
Hynix had a choice to either guarantee usage for a twenty year period of time, or pay for the
power station up front, of which they chose the latter. Hynix's actual electricity rates and usage
statistics are not public information. However, from this discussion, we assume that Hynix's
presence in the community has caused no net benefit or cost with respect to electricity usage.
Total Cost: zero
Total Benefit: $8.6 million
Hynix also requires large quantities of water. We identify three ways in which this may
cause additional costs to the community: the expansion of water filtration plants, the probability
of a drought, and community water rates. Firstly, Hynix's move into Eugene did "move up" the
expansion of an existing water filtration plant. According to Steve West, another EWEB
representative, Hynix's water rates were built to include any maintenance or expansion fees
EWEB would need to cater to their needs. These fees, like those for electricity, were built into
their water rates. According to him, this happens for all large corporations who will affect water
capacity in any way. However, the exact figure for the water filtration station's expansion is not
The greater probability that the presence of Hynix could cause emergency restrictions due
to such things as a drought is another potential cost. According to Steve West, there would need
to be a drought of at least 3 years for Hynix's water usage to become a problem for the rest of the
community. In addition, EWEB has lost three major water consumption contracts in the past few
years; HMT, Chef Francisco, and Agrapak, the last of which used more water daily than Hynix.
So EWEB concludes that Hynix actually forms an important part of their corporate revenue base,
and lack of capacity issues have been significantly reduced with the closing of these companies.
However, if one knows the cost of a drought and the probability, they can assess the "cost" the
possibility of a three year drought. We mention this in order to give credence to the fact that, if a
long drought occurred, Hynix's high water consumption could affect the residents of Eugene.
According to EWEB, large corporations are asked to voluntarily reduce their water usage in the
event of a drought, but this figure will demonstrate the cost to the county of the possibility that
Hynix could affect local water rates. However, EWEB could not provide a cost for us to use in
The problem with assessing drought probability is ascertaining the exact definition of
drought. Here we turned to one of the State Climatologists, George Taylor who operates out of
Oregon State University, for recommendations. He suggested the Surface Water Supply Index as
a good source to ascertain drought. The SWSI is maintained by the Natural Resources
Conservation Service (NRCS), a sub-department of the US Department of Agriculture. It
measures surface water availability (as coming from two sources: reservoir carryover and spring
and summer runoff) by a scale ranging from -4.1 (very dry) to 4.1 (very wet). 0.0 is the median
water supply throughout the history of the index, which dates back to 1974 (see graph A). The
average surface water supply index in summer from 1997 to 2002 is .55, with the low point
being the summer of 2001, where the index dipped to -3.2, and the high point being the summer
of 2002, where levels rose to 2.5 (see table B).12 What we can gain from this, is that historically,
since 1974, levels have not dipped below median for a constant time span of 3 years, which
implies that the probability of a 3 year drought is extremely low. The one high risk time span
appears to have been between 1985 and 1988. Here, the SWSI just barely touched 0 in 1986, and
the rest of the time it stayed below zero. If a probability could ever be assessed, someone would
simply need to multiply it by the possible water rate hikes due to Hynix, to assess the "cost" of
Hynix's presence in Eugene with regards to drought.
Finally, we come to effects of local rates. Like the electricity contract, EWEB took pains
to make sure Hynix's move would not affect local water rates. Hynix has a "ceiling" usage built
into their contact. If they exceed 1.8 gallons a day, they are assessed a larger fee, per gallon, for
any water used above that ceilingI. This is also to protect the community from any excess rates
derived from Hynix's high water usage. The total cost of Hynix's electricity and water usage and
the usage statistics are classified information. Therefore, the only cost assessed to allowing
Hynix's water and electricity consumption is that of the possibility of three year drought.
Total Cost: uncalculated
The West Eugene Enterprise Zone sits on a large acreage of wetlands. This presents
many problems for companies wanting to locate there. Firstly, they must establish there are no
endangered plants or wildlife that will be destroyed due to their facility's construction, and
secondly, they must compensate for wetland destruction by creating new wetland equal to twice
the amount they fill. This is mandated through a state law that governs Oregon wetlands, which
also created the Wetland Mitigation Bank to oversee and enforce wetland usage and regulations.
Under the law, any corporation destroying an acre of wetland must create two more to replace it.
There was some belief in the community that the Wetland Mitigation Bank may have awarded
already established wetlands to Hynix in order to stave off some of the costs associated with
Hynix was originally estimated to need 2.5 million gallons of water per day. They have never used this amount,
and according to EWEB, rarely approach the ceiling value.
building new land to compensate for the destruction of wetlands within the enterprise zone. This
According to Neil Bjorklund, a planner for the city of Eugene who was present during the
Eugene hearings regarding Hynix and the West Eugene wetlands, Hynix did in fact pay to restore
or enhance wetlands in order to compensate for wetland destruction in the required 1:2 ratio.II
Hynix destroyed 10.4 acres of wetland in order to build their facility. To compensate, they
restored 12.97 acres of wetland, restored a pond encompassing 2 acres within a wetland area,
enhanced 6.92 acres of wetland, and enhanced a stream corridor encompassing 3.7 acres within a
wetland area. This totals 25.59 acres of restored or enhanced wetland built to compensate for
10.4 destroyed acres, which fulfills the 1:2 ratio dictated by law13. What Hynix paid for the
restoration is not public information and could not be found for this study, but according to Neil
Bjorklund, it would cost the city approximately $30,000 an acre to restore wetland. He believes
Hynix may have been able to pay a slightly lower figure because they contracted privately to
restore the land, but they did also hire a consultant to oversee all these activities.
Hynix was also assessed a small fee to conduct a rare plant survey, which verified they
would not be destroying endangered plants in the construction of the plant. Although it would be
nice to have these numbers, they are not necessary for this section because by fulfilling the law,
Hynix eliminated any cost as dictated by law. We understand that placing a numerical value on
wetlands is controversial, thus we are accepting the law Oregon established as the standard for
this study, assuming that because it was created by a representative government, that it reflects
Wetland restoration, according to the City of Eugene, means reconstruction of previously destroyed wetland, while
wetland enhancement is improvement of badly functioning wetlands. They both include pollution repair, native
plant restoration, and fill dirt removal.
the preferences and opinions of the majority of the voting public. However, we respect that there
may be those who disagree with this standard of wetland evaluation.
Total Cost: zero, as the ratio requirement was fulfilled by law.
One possible cost of any large factory is congestion. With an average of 850 employees,
Hynix has the potential to significantly increase traffic in the West Eugene Area. Congestion is a
cost because it increases wait times for commuters, which implies an opportunity cost to the
commuter. Opportunity cost is assessed as the next best a person could be doing, i.e. if a person
is sitting in traffic the opportunity cost of that could be them arriving at work earlier, or watching
television. Opportunity cost is difficult to quantify, but it needs to be mentioned to thoroughly
analyze any cost situation. In this case, we could not obtain records demonstrating the increase in
traffic on the streets surrounding Hynix. We did learn however, from Denny Braud, a city
planner in Eugene, that these roads do not require above average maintenance, which suggests
that there is not currently a congestion issue. However, there may be a lag effect, where the
current congestion may not show wear and tear on the roads for a few years. Even if we did
obtain the proper records, we would have to isolate Hynix's increased traffic from other factors,
such as the fact that housing has expanded into the same area. That is a study within itself. So
although we cannot present a numerical result, we feel it is important to mention congestion as a
Total Cost: uncalculated
TOTAL NEGATIVE IMPACTS:
Tax exemptions: 46 million
City Road Enhancement: 3 million
Electricity and Water Consumption: zero
Drought Potential: uncalculated
Wetland Destruction: zero
Total Costs: 49 million dollars
Positive Economic Impacts
We now discuss the positive economic impacts to Lane County with regards to the
presence of Hynix. We have already cited some benefits as they directly applied to issues related
to costs, and those will be totaled at the end, but not included in this section (see Table A in
executive summary). Unlike many of the costs, benefits are more difficult to calculate. We will
try our best to explain our methods as we proceed, but for additional guidance please reference
the methodology section of this paper, which provides further explanation.
Wages and Benefits
The most obvious positive local impact generated by a new plant in a community is the
wages and profits it generates for employees and owners. Given that Hynix is owned by a
foreign parent company, profits accrue to owners abroad. Thus, the positive economic impact
stems primarily from wages and benefits paid to local employees. From 1997 to 2003, Hynix
paid an average wage plus benefits package of roughly $55,649.50 per year, of which the
average wage-only portion of the income package paid at Hynix is $35,615.68, which is 64% of
the original figure.14 Based on standard applied payroll taxes, we can estimate that payroll taxes
average $3650.61 per employee, while benefits, including health insurance, 401k contributions,
and paid time off are $16,383.72. Thus, the average wage and benefits for Hynix employee
during our focus period is estimated to be $51,998.89 (excluding payroll taxes). From numbers
given to us by Jerry Olson at Hynix, the average number of employees at Hynix from 1997
through 2001 was 644. This means that Hynix's total employee compensation from 1997 to
2001 was $167,436,426. However, the amount of additional local spending these wages and
benefits create is likely greater than this amount, as it has what is termed a “multiplier effect” on
local community spending.
We will use an example to illustrate how the multiplier effect works. Suppose someone
is paid $100 a week. This initial $100 not only benefits the community, but generates additional
local spending because the person or firm receiving the $100 will use some of the money to buy
goods and services locally, thus providing income to local producers. These producers, in turn,
use some of this income to pay wages to employees and pay for supplies from other local
producers. These local wage earners and suppliers then use some of this income to buy goods
and services locally, etc. Thus, the initial $100 generates a domino effect of re-spending through
the local economy which generates more local spending than the initial $100. This is the
multiplier effect. It tells us that spending in the local economy is "multiplied" out through a
series of purchasers who spend a portion of the initial income locally at every step in the series.
The IMPLAN model discussed earlier calculates multipliers for various sectors of the economy.
We have chosen to use the IMPLAN model's calculation for Lane County's multiplier in the
semiconductor industry. In particular, the multiplier we use is IMPLAN’s "total value added"
multiplier which is used to determine how local value added spending will be multiplied in the
community. This multiplier is appropriate since wages and benefits are typically the main
portion of value added for any firm. The IMPLAN value added multiplier for the
semiconductor industry in Lane county is 1.52. This means for each dollar paid to an employee
of Hynix in wages and benefits, $1.52 of actual local spending will be generated in the Lane
county. Based on the above calculation of wages, this means the total positive impact of wages
and benefits paid by Hynix is:
51998.89 * 644 (average number of employees) = $33,487,285.16 * 5 (beginning of 1997 to
beginning of 2002) = $167,436,425.80
1.52 (multiplier) * $167,436,425.80 = $254,503,367.20
Note that we are calculating for the years from the beginning of 1997 to the beginning of 2002,
as those are the years for which we have data. Thus, in summary, the estimated total positive
impact from wage and benefits to the Eugene-Springfield area is $254,503,367.20.
This above calculation is the standard methodology in the literature to estimate the
additional local spending generated in a community by the presence/addition of a new firm in a
community. This calculation, however, assumes that the new plant’s employee positions are
additional ones that the community would not enjoy otherwise. Alternatively, one could make
the assumption that the number of jobs in the community does not change at all, as workers come
from previously existing firms and some marginal firms drop out due to competition (and higher
wages) for labor. This is likely an extreme assumption, particularly with Eugene-Springfield’s
currently high unemployment rates, but nevertheless gives us a lower bound on the positive
impacts from Hynix’s presence. In fact, even under this restrictive scenario, Hynix brings
increased local spending due to the fact that they are compensating their employees at a much
higher level than the average comparable worker in the area.
In particular, assume that without Hynix, its current employees would be making the
average wage for production workers in Eugene-Springfield, which is $26,760 according the
2001 data from the Bureau of Labor Statistics.14 Taking the difference between Hynix’s average
wage and the city’s average producer wage, multiplying it by the number of employees, the
numbers of years, and the multiplier yields the following calculations and results.
$35,615.68 - $26,760 = $8,855.68 * 644 = $5,703,057.92 * 5 = $28,515,289.60 * 1.52 =
Thus, under this scenario, the additional local spending generated by Hynix from additional
wages is $43,343,240.19. This calculation only involves the wage component, assuming that
existing firms benefits packages are identical to Hynix’s even though they do not pay as high of
wages. If we assume that the benefits offered by Hynix to its employees are 33% higher than the
local norm, as is true with wages, we get a calculation of $70,069.228.33 of additional local
spending due to increased wages and benefits provided by Hynix. Either way this scenario is
fairly restrictive in its assumptions. Yet, our overall net impact of Hynix (once all the negative
and positive impacts) are added up, would still yield a positive net impact under this very
However, one final proviso is that $100 in wages is likely not the same as an outright
benefit of $100, since the person must work for the wage income. Economists typically think
that people suffer disutility (a type of cost) from working. On the other hand, studies of
unemployed persons suggest that a job has many positive impacts on the person self-esteem and
personal self worth. This is particularly appropriate to consider if some of the Hynix workers
would be unemployed without the presence of Hynix. These issues are hardly ever considered in
typical business development impact studies, but we mention them here for completeness.
Tax Revenue Increases
Whenever someone receives a higher wage than they previously did, or an unemployed
person gains a wage at all, the tax revenue base is affected. This too can be a source of positive
impact on the local economy. The first, most obvious outcome is that there will be more taxes
paid to the state. The other is more indirect: one theory as to why enterprise zone tax exemptions
will not negatively affect the tax base is because the workers employed by those corporations
will receive higher than average pay, thus allowing them to upgrade their property. In essence,
these new jobs allow people to either buy property or move into more quality rental property,
both in turn increasing the amount of property taxes paid to the state.
In order to estimate the amount of taxes paid by Hynix employees, we take the average
taxes paid by a Lane county resident, $2,53515, and multiply it by 644 employees, the average
number of employees Hynix maintained from 1997-2001. Using these two figures, we can find
that the average taxes paid are 2535*644 = $1,632,540. We multiply this by 5 to account for all
the years these employees have been paying this income taxIII. 5*1,632,540 = $8,162,700. Now
we face a problem. The tax code changes every year and its difficult to estimate how much
money paid in state taxes actually returns to Lane County. We have decided to estimate that,
since 13% of Oregon's GDP comes from Lane County, that approximately 13% of Oregon taxes
will also return here. This is an inexact estimate, and if one wanted to create a better one, they
could examine the annual tax codes for the last five years to determine exactly how much tax
money returned to Lane County. (.13)(8162540)= $1,061,130.2.
Why we don't simply find the tax bracket for the average income paid by Hynix and derive the taxes from there?
Due to the non-linear fashion of the tax code, and the fact that we do not have data concerning what deductions or
credits, this would not be a very accurate estimate of taxes paid. This technique is commonly used in economic
It is harder to calculate the possible contribution of increased property tax. We do not
have data on how many of Hynix's employees own homes, so again we must rely on averages.
According to the Oregon Economic and Community Development Department 68% of
Oregonians owned their own homes as of 199916. If we assume, as Hynix states, that they
maintain an average of 644 employees, at least 70% being hired from Lane County, we can
conservatively hypothesis that (.7)(644) = 451 employees live in Lane County, and (.68)(451) =
307 of them may own their own homes. Thus, if we take the average property taxes paid in the
state of Oregon, which was $2017 for urban areas and $956 for rural areas in 1999, we can
deduce an upper and lower bound for the property taxes possibly paid by Hynix employees.
2017*307 = $619,219*5 = 3,096,095. (calculating for the time span between 1997 and 2002),
which would be the upper bound assuming all home owners pay the higher property tax, and
956*307 = 293,492*5 = $1,467,460 would be the lower bound, assuming all home owners pay
the lower property tax.
Total Tax Benefits:
Income Tax: $1,061,130.2
Property Tax: Between $1,467,460 and $4,096,095
TOTAL = Between $2,528,590.2 and $5,157,225.50
Under the more restrictive assumption that Hynix does not lead to any additional
employees in the area, only higher wages, the numbers will be much smaller. In terms of payroll
taxes, we can estimate that they will be 33% higher for Hynix employees because wages are 33%
higher than the county norm for production workers. This means the following calculation and
It is hard to estimate how much higher incomes affect average home ownership among
employees and thus, lead to increase in property taxes. Thus, to be conservative we assume no
increase in property taxes under the conservative scenario.
Short-run benefits from initial plant construction are often a large focus of studies
examining the economic impact of business development. The largest disappointment in our
study is the inability to find a figure representing how much money Hynix put into the local
community during their facility's construction. It is common knowledge that Hynix spent
upwards of 1.2 billion dollars to build the Hynix factory, but how much of that was spent in Lane
County remains a mystery. Hynix either did not keep records regarding that, or was unable to
release the information to us at the time of this study's publication. We made various calls to
construction firms around town, who recalled being subcontracted, but could not provide
information or records specifying numbers of employees, man hours, or pay. There is no good
way to estimate how much money was spent in Oregon, because we don't even know how much
of this figure was spent in the United States. Suffice it to say, there is probably some significant
amount of money that is lacking from the benefit section of this study in regards to the plant's
Total Positive Economic Impact
Summing up we have the following accounting of the positive economics impacts from
Hynix’s presence in the Eugene-Springfield area from 1997-2001. The first set of numbers on
the next page follows the standard methods of calculating these impacts, which typically assume
that all employees in the plant are additional jobs. The second set of numbers makes the very
conservative assumption that additional jobs are in the local community due to the plant, but
recognizes that Hynix is paying higher wages and benefits. Note that some of the positive
economic impacts were discussed in the negative economic impact sections.
Total Positive Economic Impact – Standard Methods:
Total Wage and Benefits with Multiplier Effect: $254.5 million
Taxes from Employees: Between $2.5 million and $5.2 million
Tax Payments from Hynix: 5.2 million
Hynix Payments to Schools, City, and County in 2002: $0.5 million
Electrical Substation: $8.6 million
Systems Development Charge: $3.7 million
TOTAL POSITIVE ECONOMIC IMPACT: $275.0 million - $277.7 million
Total Positive Economic Impact – Conservative Methods:
Total Wage and Benefits with Multiplier Effect: $70.1 million
Additional Taxes from Employees: $0.4 million
Tax Payments from Hynix: 5.2 million
Hynix Payments to Schools, City, and County in 2002: $0.5 million
Electrical Substation: $8.6 million
Systems Development Charge: $3.7 million
TOTAL POSITIVE ECONOMIC IMPACT: $88.5 million
According to our calculations (summarized in the executive summary), when one sums
up the negative and positive economic impacts on Lane County from Hynix’s presence in the
community, the net impact has been quite positive, totaling approximately $225 million using
standard methodologies. Even under very conservative assumptions, we estimate a positive net
impact of almost $40 million ($88.5 million in positive economic impact minus $49 million in
negative economic impact). As we have noted, we have not been able to include estimates of all
possible economic impacts in our calculations, including impacts from initial construction,
congestion effects, etc. However, we have been able to estimate most impacts commonly cited
as important and hope the calculations provide a good framework with which to analyze the true
economic impact of Hynix on Lane County. Based on our data, and the fact that our largest
missing piece of data, construction spending, would all fall under the benefit category, it is our
analysis that Lane County made the correct choice to offer development incentives to Hynix.
Hopefully more academic literature will be published in the future regarding methods of
calculating benefits, at which point someone could re-examine this study’s methods, and draw an
even more precise conclusion. Although we have discovered there is a positive net impact, it is
important to point out some factors, the first being that this study has only evaluated the past.
Our data regarded salaries, wages, and tax base changes due to Hynix employees is only through
the end of the 2001 fiscal year, and our negative impact data is through 2002. We have not
forecasted any future benefit or costs related to Hynix and cannot guarantee that the current
trends of positive net impacts will continue.
One of the key arguments against Hynix in the media throughout the past 6 years stated
that encouraging small business development would be better for Eugene in the long run than
inviting a large scale company that could easily leave town when the economy turned south. A
comparison paper evaluating different scenarios, such as comparing the effects of one large
business to several smaller businesses, would be an interesting extension of this project. Another
interesting comparison would be between Hynix and one of the other large companies residing in
a Lane County enterprise zone, such as Monaco, Marathon, or Sony, the last of which just
It is also important to recognize the long term effects of the Eugene community's reaction
to Hynix. In comparison with other large scale development incentive programs, which are
always controversial, we feel Eugene's community may have demonstrated a more negative
reaction than most, or perhaps its reaction was more clearly publicized. This could possibly
discourage other large businesses from locating to Eugene.
Finally, the Hynix's effect on the housing and property values in its areas as an evaluation
beyond the scope of this study. Most likely, when Hynix first arrived, housing values increased,
but during periods of retooling or major change (for example, when Hyundai and Hynix split into
two different companies) values may have decreased. However, there are other factors, such as
the economy, which make Hynix's effect on the market difficult to isolate. But the effect of large
companies such as Hynix on the housing market is another valuable issue that could be
With all the controversy and miscommunication in the community, we hope that this
compilation and analysis of information regarding what has happened thus far with Hynix can
serve as a guide and a tool to help Lane County better understand the effects of development
incentives. While the majority of studies find inconclusive or negative economic impacts (not
benefits) in regard to development incentives, we feel the fact that such a large net positive
impact was discovered is significant. We attribute the uniqueness of our situation to the fact that
the city and EWEB managed to minimize their costs in comparison with other development
programs. For example, when Kentucky offered Toyota financial backing to open a factory there,
they gave them outright cash gifts, and allowed many city organizations to make deals with them
in order to minimize the firm's costs. In Eugene, EWEB made it clear that Hynix would pay for
any and all new facilities they needed to accommodate their usage, and pay extra for independent
contracts to cover their electricity. We suspect strong moves like this one and others were the
cost minimizing factors that created a positive economic impact for Eugene-Springfield.
Hopefully, by examining what happened here in the last 5 years, we can discover all the reasons
why this situation was different, and use that to Lane County's economic benefit in the future.
Table B: SWSI chart for summers 1997-2002
Year June July August Average
2002 1.8 2.2 2.5 2.167
2001 -3.2 -3.2 -3.2 -3.2
2000 0 -0.1 -0.2 -0.1
1999 1.6 1.5 1 1.367
1998 1.1 1 1.2 1.1
1997 1.2 1.3 2 1.5
mean 0.42 0.45 0.55 0.55
Graph A: Willamette Basin – Historical SWSI – 5 Month Averages
Cassell, Jonathan. "Hyundai to Break Ground with US Fab." Electronic Buyers News. 29 May 1995. Online. Lexis
Nexis. 3 December 1999.
Poole, Kenneth, George Erickcek, Donald Iannone, Nancy McCrea, and Pofen Salem. "Evaluating Business
Development Incentives." Accra. August 1999, p. 36-37.
Kidd, Joe. "Hyundai Won't Need Hearings." Register Guard. 1 June 1995, p. 1C-2C.
Calatone and Loof. The Economic Significance of Toyota Motor Manufacturing, U.S.A., INC, In Kentucky. 1992.
Singell, Larry. "A Study of the Economic Impact of the University of Oregon." 2002.
Ayre, Arthur. "The Economic Impacts of a Steel Mill in Coos County." Oregon Economic Development
Department. 7 August 1997.
Minnesota IMPLAN Group. IMPLAN informational site. 2003. <http://www.implan.com/default.htm>
Braud, Denny. Personal interview. 6 May 2003.
Olsen, Jerod, Hynix representative. Personal email. 14 May 2003.
McDonald, Sherri. Personal interview. 5 May 2003.
Hynix. Commitment to the Community. 18 June 2001.
NRCS. SWSI Tables. June 2003. <http://www.or.nrcs.usda.gov/snow/watersupply/swsi.html>
Bjorklund, Neil. Personal interview. 7 May 2003.
Olsen, Jerod, Hynix representative. Personal email. 14 May 2003.
US Bureau of Labor Statistics. 2001 Metropolitan Area Occupational Employment and Wage Estimates
for Eugene-Springfield, OR MSA <http://www.bls.gov/oes/2001/oes_2400.htm#b00-0000>
Oregon Dept of Revenue. Oregon Dept of Revenue Statistics Report for 2000. 2001.
Oregon Economic and Community Development Department. 2001 Annual Report. 2001.